The Association between Drug-Related Problems and Length of Stay of Type 2 Diabetes Mellitus Patients

Introduction Drug-related problems (DRPs) are treatment-related occurrences that affect therapeutic efficacy. In a previous study, approximately 279 out of 330 (84.5%) patients with type 2 diabetes mellitus (T2DM) had experienced at least one DRP, including non-optimal drug effects (n = 240, 52.7%) and indications without medication (n = 137, 30.1%). Patients who were hospitalised for 5–10 days had the highest number of DRPs. Therefore, this study investigates the association between DRPs and length of stay (LoS) in patients with T2DM. Methods A cross-sectional study was conducted from January 2020 to May 2023 at Rumah Sakit Akademik, Universitas Gadjah Mada, Yogyakarta, Indonesia. Clinical pharmacists reviewed electronic health data to examine DRPs. The Fisher’s exact test evaluated the association between DRPs and LoS. Results A total of 60.7% (n = 17) of the participants were females, with the majority falling into the age group ≥ 65 years old (n = 11, 29.7%). A significant portion experienced LoS > 7 days (n = 17, 60.7%). Antidiabetic monotherapy was predominant, and the categories of DRPs included adverse drug reaction (n = 15, 40.5%), dosage too high (n = 6, 16.2%), wrong drug (n = 6, 16.2%), non-adherence (n = 4, 10.8%), need for additional therapy (n = 4, 10.8%) and dosage too low (n = 2, 5.4%). A significant association was observed between non-adherence and LoS (P = 0.016). The possibility of experiencing LoS of 1–7 days increased by 3.43 times with improved non-adherence (OR = 3.43; 95% CI: 1.83, 6.39). In this context, non-adherence refers to DRPs associated with the non-compliance of patients with the prescribed treatment plan. Conclusion This study concludes that non-adherence was significantly associated with hospital LoS.

According to the International Diabetes Federation (IDF), the global prevalence of DM reached 463 million in 2019, with Indonesia ranking 7th with 10.6 million patients.The worldwide prevalence is expected to increase to 578 and 700 million in 2030 and 2045, respectively, marking a 51% rise (8).In Indonesia, the death rate due to diabetes ranks 2nd after Sri Lanka (9).As per the Basic Health Research/Riset Kesehatan Dasar (RISKESDAS) (10), Yogyakarta Special Region Province ranked 3rd for diagnosed DM cases.Treatment modalities include oral antidiabetic drugs, insulin injections or a combination of both, depending on the doctor's diagnosis (10).
Movva et al. (11) observed that patients with length of stay (LoS) ranging 5 days-10 days had the highest incidence of DRPs.These problems contribute to increased treatment costs and impact the LoS due to therapeutic inefficacy (12).Previous reports investigated the influence of DRPs on LoS in patients with hypertension in pregnancy, revealing a significant increase (13).The present study aimed to determine the association between DRPs and LoS in patients with T2DM.

Study Design, Location and Duration
A cross-sectional study was conducted at Rumah Sakit Akademik, Universitas Gadjah Mada, Yogyakarta, Indonesia, from January 2020 to May 2023 (since clinical pharmacists do integrated documentation in electronic health record or EHR).

Study Sample and Patient Selection
The study comprised all patients diagnosed with T2DM.The inclusion criteria comprised individuals ii) aged 18 years old and above, ii) patients with T2DM as a primary diagnosis, iii) patients with blood glucose tests, and iv) patients who were actively prescribed antidiabetic medication.Those with incomplete records or missing data were excluded.Determining T2DM as the primary diagnosis relied on International Classification of Disease, 10th Revision (ICD-10) codes assigned by the medical records department.

Data Collection and Identification of DRPs
Clinical and demographic variables, such as gender, age in years, LoS and type of medications used, were collected along with comorbidity information.This study examined DRPs using electronic health data reviewed by clinical pharmacists.Upon hospitalisation, the clinic's pharmacist determined the presence or absence of DRPs.During hospitalisation, the presence of DRPs was assessed and the identification results were recorded on the integrated patient progress record sheet within the EHR.
Data related to DRPs were collected by the clinical pharmacists, including current issues, potential occurrences and recommendations offered.The data was extracted and classified into the DRPs category using the classification system of Cipolle, which comprises the following: need for additional therapy, unnecessary therapy, wrong drug, dosage too low, ADR, dose too high and non-adherence (2).The term non-adherence refers to DRPs linked to noncompliance of patients with the prescribed treatment plan.

Statistical Analysis
Descriptive statistics were used for data analysis, presenting patients' characteristics and DRPs as numbers and percentages.The Fisher's exact test evaluated the association between DRPs and LoS.The odds ratio (OR) quantified the strength of the association, with an OR < 1.00 suggesting a reduced probability of LoS of 1-7 days due to DRPs.OR = 1 showed no association, while an OR > 1 suggested a higher probability of LoS of 1-7 days.The significance of this association was determined by a P < 0.05, with a 95% confidence interval (CI).Data analysis was performed using the SPSS version 12.0.

Demographic Characteristics
A total of 203 patients with T2DM were initially included but only 28 patients were identified to have at least one DRP and were subsequently analysed (Figure 1).Table 1 shows that out of these 28 patients, 39.3% were males (n = 11) and 60.7% were females (n = 17).The majority of patients were in the age group ≥ 65 years old (n = 11, 29.7%), followed by the age groups of 55 years old-64 years old (n = 9, 24.3%), 45 years old-54 years old (n = 5, 13.5%) and 18 years old-44 years old (n = 3, 8.1%).In terms of their LoS, the majority of patients were hospitalised for more than 7 days (n = 17, 60.7%).Additionally, most patients received antidiabetic monotherapy (n = 20, 71.4%).The prevalent additional health problems included three comorbidities (n = 9, 32.1%).Table 3 represents the details discussed above.

Factors that were Significantly Associated inT2DM with Drug-Related Problems
Table 5 shows a significant association between non-adherence and LoS (P = 0.016).The possibility of experiencing LoS of 1-7 days increased by 3.43 times with improved nonadherence (OR = 3.43; 95% CI: 1.83, 6.39).

Discussion
The occurrence of DRPs has been associated with prolonged hospital stay, increased financial burden, and nearly a two-fold higher risk of mortality (14).Due to the considerable health and financial costs, hospitalisation due to DRPs is a significant concern to both patients and healthcare providers (15).Maximising medication efficiency and preventing these problems is crucial for improving healthcare, reducing expenses and potentially saving lives (16).DRPs are assumed to be expensive, serious and complicated issues for the healthcare system, often associated with polypharmacy, multimorbidity and advancing age in diabetic patients.Risk factors for these problems in diabetic patients also include renal impairment, inadequate cholesterol management, cardiovascular disease and LoS (17).
DRP development was significantly associated with females.The World Health Organization (WHO) (18) offered a compelling explanation, suggesting that females are more prone to being overweight, obese and physically inactive.Additionally, the higher occurrence of poor glycaemic control in females may be linked to biological and psychosocial factors (19,20).Despite several reports suggesting that older age (> 60 years old) poses a risk for DRPs (21), this study did not find a statistically significant association.The increased occurrence of these problems in geriatric patients might be explained by the correlation between lower creatinine clearance (CrCL), high polypharmacy and a higher number of DRPs (22).Aging is generally associated with an increased risk of ADR and other related issues due to slowed metabolism and excretory processes.Numerous studies have stated that elderly patients taking multiple medications are more susceptible to DRPs (23).
This study shares similarities with Bathari et al. (24), who showed rapid-acting insulin was the most commonly prescribed antidiabetic.Insulin is the ideal option for precise blood glucose regulation, allowing for prompt adjustments based on glucose readings.To prevent hypoglycaemia, insulin therapy generally begins with a small dose.This may include oral medication in addition to insulin or an insulin combination therapy (25).
Consistent with other studies, drug classes frequently linked to DRPs included gastrointestinal, endocrine and cardiovascular medications (26).According to investigations conducted in the UK, Saudi Arabia (27) and Wolaita Sodo, Ethiopia (28), a higher number of comorbidities was linked to an elevated risk of developing at least one DRP.This relationship may arise from individuals with more comorbidities being more prone to taking multiple medications, which may lead to nonadherence and an increased susceptibility to ADR.
A study by Sharma et al (29) revealed a significant association between comorbidities and DRPs.The use of multiple drugs can result in drug-drug interactions and a complex medication schedule.The frequent administration of medication and an increased number of pills may contribute to the occurrence of DRPs.
In this study, ADRs were the most prevalent DRPs.The possibility of encountering these problems increased due to multiple therapies, particularly when six or more different types of medications were involved.This showed the importance of preventing and managing drug interactions (30).Patients with comorbidities, particularly those receiving seven or more medications (polypharmacy), were found to be at a higher risk of ADRs and drug interactions (31).
Studies conducted in Northern Sweden stated that patients admitted to hospitals without clinical pharmacist services frequently experienced inappropriate drug use and drugdrug interactions (32).Drug interactions and dosage issues were identified as frequent DRPs (33).Consistent with a German study (34), where inappropriate medication use and non-adherence problems were also prevalent.Additionally, the LoS was associated with potential drug interactions (35).
DRPs have a significant impact on the quality of life of hospitalised patients, resulting in prolonged hospital stays, higher healthcare expenses and mortality (36).According to a retrospective analysis, DRPs related to nonadherence and adverse effects had the highest potential for clinical significance and risk of harm (37).Additionally, a systematic review and meta-analysis identified a significant relationship between poor medication adherence and the incidence of these problems (38).
Patients experiencing ADR and drug interactions, as shown by Kurniawati et al. (39), tend to have longer LoS.An extended LoS not only requires more hospital resources but also increases the costs of treatments (40).Additionally, another study showed that longer hospitalisations were more prone to have at least one type of DRP, potentially due to an increased risk of nosocomial infections requiring extensive therapeutic interventions (41).Non-adherence issues further compound the challenge, hindering the achievement of therapeutic goals and potentially leading to prolonged LoS.
Increased LoS manifests the challenges faced by inpatients due to medical intervention (42).Hospital pharmacists play a crucial role in mitigating this negative impact, although this task poses a significant challenge, given the demanding workload in hospitals (43).This implied that emphasis should be placed on prioritising patients at the highest risk and in need of guidance (44).Several independent factors contribute to medication-related harm during hospitalisation, including advanced age, comorbidities, impaired kidney function and the use of high-risk medications (45).In addition to taking more prescription drugs, patients with multiple chronic conditions are more to experience medication-related issues.In older adults, physiological changes associated with aging affect drug pharmacokinetics and pharmacodynamics, leading to a higher risk of adverse medication events.Additionally, comorbidities and advanced age can contribute to polypharmacy and an increased risk of non-adherence (46).Table 5 shows patients without compliance problems had a longer LoS (> 7 days).Pharmacists agreed that the risk of experiencing an adverse medication event increased with duration of hospitalisation.Consequently, LoS is one of the priority criteria for clinical pharmacist therapy monitoring.Pharmacist interventions improved medication adherence in most studies.The interventions by pharmacists include the evaluation of medications, delivering educational sessions and counseling, and offering therapeutic suggestions to prescribing healthcare professionals (47)(48)(49).Additionally, patients with longer LoS were stated to have a lower possibility of medication changes.To achieve optimal outcomes and enhance medication safety, it is essential to promptly identify high-risk cases and proactively prevent or minimise drug-related issues (44).
An insightful study conducted in Pakistan emphasised the importance of keeping clinical knowledge of pharmacists up-to-date to improve their ability to make interventions that effectively lower the incidence of DRPs (50).In the future, modern technology can be leveraged to both prevent and identify DRPs, contributing to enhanced safety of patients and the achievement of therapeutic goals.

Strengths and Limitations of the Study
The strength of this study lies in the identification of DRPs using the classification system of Cipolle, a recognised method employed by clinical pharmacists.However, the study is limited by its small sample size.Additionally, the performance of pharmacist intervention was not evaluated.Future research requires a more extensive confirmatory and multicentre study to overcome these limitations.

Conclusion
In conclusion, ADRs were the majority of DRPs identified in this study.Additionally, a significant association between non-adherence and extended hospital LoS was observed.Pharmacists play an essential role in influencing the outcomes of T2DM patient management by proactively preventing DRPs through pharmaceutical care.

Figure 1 .
Figure 1.Flowchart of the patient selection

Table 1 .
Demographic characteristics of the patients Note: *comorbidity

Table 2 .
Distribution of the type of prescribed antidiabetic

Table 3 .
Distribution of the type of comorbidities (continued on next page)

Table 4 .
Distribution of the classification of DRPs

Table 5 .
The associate between DRPs and LoS in T2DM patients 37. Westberg SM, Derr SK, Weinhandl ED, Adam TJ, Brummel AR, Lahti J, et al.Drug therapy problems identified by pharmacists through